# InferX

Free Developer Tier · API key · code · Last updated: 08 Nov 2026

12 free models on InferX including Qwen3.6 35B A3B FP8, deepseek-v4-flash, and Ornith-1.0-35B-FP8. OpenAI-compatible endpoint (https://model.inferx.net/endpoints/v1), up to 256K context. Last updated: 08 Nov 2026.

The Grant: InferX provides zero-cost inference endpoints for cutting-edge open models including Qwen 3.6, Gemma 4, and DeepSeek variants.
The Output: Ultra-low latency OpenAI-compatible completions with streaming and structured output support.
The Process: Sign in to generate an API key and query endpoint URLs directly.
The Condition: Rate-limited for community development and prototyping.

Source: https://model.inferx.net/endpoints · Last updated: 08 Nov 2026

| Model | Cost | Context | RPM | TPM | RPD | Day tokens | Verified |
| --- | --- | --- | --- | --- | --- | --- | --- |
| Qwen3.6 35B A3B FP8 | $0 | 255.9K | 60 | Not published | 2,000 | Not published | 2026-11-08 |
| deepseek-v4-flash | $0 | 1.0M | 60 | Not published | 2,000 | Not published | 2026-11-08 |
| Ornith-1.0-35B-FP8 | $0 | 255.9K | 60 | Not published | 2,000 | Not published | 2026-11-08 |
| Qwen3.6 27B FP8 | $0 | 256K | 60 | Not published | 2,000 | Not published | 2026-11-08 |
| mimo-v25 | $0 | 1.0M | 60 | Not published | 2,000 | Not published | 2026-11-08 |
| Qwen3 Coder Next FP8 | $0 | 250.1K | 60 | Not published | 2,000 | Not published | 2026-11-08 |
| Agents-A1 | $0 | 255.9K | 60 | Not published | 2,000 | Not published | 2026-11-08 |
| Devstral-2-123B-Instruct-2512-int4-AutoRound | $0 | 125K | 60 | Not published | 2,000 | Not published | 2026-11-08 |
| Qwen3-Embedding-8B | $0 | 32K | 60 | Not published | 2,000 | Not published | 2026-11-08 |
| Qwen3.6-35B-A3B-fp8-no-thinking | $0 | 255.9K | 60 | Not published | 2,000 | Not published | 2026-11-08 |
| Gemma 4 31B IT FP8 | $0 | 256K | 60 | Not published | 2,000 | Not published | 2026-11-08 |
| Qwen3-Coder-Next-FP8-no-thinking | $0 | 253.9K | 60 | Not published | 2,000 | Not published | 2026-11-08 |

Machine-readable: /data/inferx.json

Numbers cross-checked against official docs and third-party trackers as of 2026-11-08. Limits change without notice — this is a map, not a contract.